Infection with HIV-1 leads to a severe depletion of peripheral CD4 T lymphocytes. This decrease is associated with phenotypic alterations of lymphocyte subpopulations  and progressive impairment of functional activity . Highly active antiretroviral therapy (HAART) including one protease inhibitor plus two nucleoside reverse transcriptase inhibitors has been capable of potently suppressing viral replication below detection limits for at least 1 year in the majority of patients who remain under treatment . Decrease of HIV-1 plasma viraemia is generally accompanied by clinical improvement and disease stabilization, even in drug-experienced subjects . Effective therapy causes an early and consistent increase in the number of circulating CD4 T cells even in patients starting HAART at an advanced stage of the disease .
Analysis of naive and memory T-cell subsets in patients responding to HAART reveals that an increase in memory cells is evident from the first months of therapy, whereas the number of naive cells is not modified  or increases more slowly later compared with the baseline number of circulating CD4 T cells . Persistence of naive and memory T cell levels is observed also in patients with rebounds in viral load . Moreover, subjects responding to HAART show reduced CD8 T-cell activation [5,7].
However, the degree of immune suppression susceptible to restoration, the extent of immune function recovery and its duration, are still controversial topics. Some authors have shown that restoration of T- lymphocyte function may only be identified regarding immune responses present before therapy . In contrast, others have demonstrated that this restoration depends on the amplitude and duration of viral load reduction and the increase of memory CD4 T cells . These studies were performed in small groups of patients with a follow-up of 1 year. In addition, the baseline mean value of CD4 T cells was higher than 100¥106/l, with the majority of patients showing CD4 T-cell counts of 100-300¥106/l [5,7,9].
Our aim was to assess whether these modifications could also be obtained in patients with AIDS and CD4 T-cell counts below 50¥106/l and to verify if the immune response may be restored against other antigens at present not investigated by others.
The effects of HAART were studied in a cohort of patients who, before HAART, showed high viral load and a loss of response to mitogens and recall antigens. During 24 months of follow-up lymphoproliferative responses (LPR) to mitogens, recall antigens and HIV-1 antigens were evaluated. Furthermore, changes in key surface molecules on T cells and plasma HIV-1 viral load were assessed and the clinical outcome of patients after treatment described.
Materials and methods
Twenty-one AIDS patients from the Department of Allergy and Clinical Immunology at University of Rome ‚La Sapienza‚ were enrolled in an open prospective study with protease inhibitor (indinavir, Merk Sharp & Dohme Ltd. Hoddesdon, UK and ritonavir, Abbott Laboratories Ltd. Queenborough, UK). Of the 21 patients, 14 were male. The median age at entry was 36 years (range, 27-48 years). According to risk factors, the patients were subdivided as follows: 11 injecting drug users, six homo/bisexuals and four heterosexuals. According to Centers for Disease Control classification , all patients belonged to the C3 group. The mean CD4 T-cell count on entry was 20¥106/l (range, 2-44¥106/l) and mean plasma HIV-1 RNA was 5.05 log10 copies/ml (range 4.0-5.98 log10 copies/ml). All subjects were nucleoside reverse transcriptase inhibitors-experienced (mean previous therapy with nucleoside reverse transcriptase inhibitors was 3.4 years) and protease inhibitor-naive. No concomitant active opportunistic infections were present at enrollment. Patient follow-up was 24 months. Karnofsky‚s score, clinical signs and symptoms, HIV-related and AIDS-defining events were monitored monthly; delayed-type hypersensitivity was evaluated on entry and every 12 months. Blood samples were taken on entry and every month for haematology, biochemistry and liver function tests, and on entry, at months 4, 12, and 24 for immunologic and virologic investigation. All subjects gave informed written consent according to the Rome University Ethical Committee procedures.
Lymphocyte phenotype was determined in whole blood samples by immunofluorescence with monoclonal antibody (MAb) combinations CD3-fluoresceine isothiocyanate (FITC), CD28-phycoerythrin (PE), CD4-peridin-chlorophyll-protein (PerCP), CD45RA- FITC, CD45R0-PE, HLA-DR-PE (Becton-Dickinson, San Josè, California, USA), CD3-FITC, CD4-PE (Ortho Diagnostic Systems, Raritan, New Jersey, USA), CD45R0-FITC (Dako, Glostrup, Denmark), CD8-tricolor (Caltag Laboratories, San Francisco, California, USA). Lymphocyte populations were quantitated by flow cytometry according to standard techniques, using a Cytoron Absolute (Ortho Diagnostic Systems).
Peripheral blood mononuclear cells (PBMC) were isolated from heparinized venous blood by density gradient centrifugation on Lymphoprep (Nycomed AS, Oslo, Norway) at 800g for 30 min. Separated PBMC (1¥105 cells/ml) were cultured in 200μl of RPMI 1640 with 10% human AB serum negative for anti-HIV-1 antibodies (N.A.B.I., Miami, Florida, USA) in 96-well flat-bottom cell culture plates (Falcon, Lincoln Park, New York, USA). CD3 MAb (OKT3, Ortho Diagnostic Systems) and phytohemagglutinin (Murex Diagnostics, Dartford, UK) at a final concentration of 25ng/ml and 1μg/ml respectively were added. Cultures were set up in triplicate.
Cultures were set up essentially as described before for mitogen cultures with the following differences: 1¥106 PBMC were seeded in each well in 100ml of RPMI 1640 with 10% foetal calf serum added in the presence of recombinant glycoprotein (rgp) 160 (American Bio-Technologies Inc., Cambridge, Massachusetts, USA) at a concentration of 2μg/ml . To test proliferative response to recall antigens Candida albicans mannoprotein  and tetanus toxoid (Wyeth labs., Marietta, Pennsylvania, USA) were added to the cultures at a concentration of 50μg/ml and 1Lf/ml respectively. Cultures were set up in triplicate.
Evaluation of LPR
After 3 days of culture at 37°C in a humidified 5% CO2 atmosphere (6 days for cultures with mitogens), 0.5μCi of tritiated thymidine (specific activity 25Ci/mmol; Amersham, Beckenham, UK) were added to each well. After further 24h (4h for cultures with mitogens) cells were harvested onto glass fibre filters by means of a cell harvester and incorporated radioactivity was measured in a scintillation counter and expressed as counts per minute (c.p.m.). Stimulation index (SI) was calculated by dividing c.p.m. of stimulated cultures by c.p.m. of the unstimulated ones. We considered as non-defective an anti-CD3- and phytohemagglutinin- driven proliferation with a SI>15.6 and >25 respectively. LPR to recall antigens were considered to be positive if SI>5. These cut-off values are 10% of the HIV-1 seronegative controls SI average. If patients were positive at baseline, a two fold increase in SI was considered at least as a significant amelioration. LPR to viral antigens were considered positive if SI>2. This cut-off value was chosen by considering that the mean ±2SD of SI obtained in cultures of PBMC from HIV-1 seronegative controls for each HIV-1 antigen tested was always <2. The same cut-off has been used by other authors in similar assays with HIV-1 antigens .
Plasma HIV-1 RNA determination
Plasma HIV-1 RNA load was measured by RT-PCR (Amplicor HIV-1 Monitor, Roche Diagnostic Systems Inc., Branchburg, New Jersey, USA). HIV-1 viral RNA copy number was calculated on the basis of the manufacturer‚s reference standards. The threshold of detection for 30 amplification cycles is 200 HIV-1 RNA copies/ml (2.3 log10).
Delayed type hypersensitivity reaction
Reactions to intradermal administration of recall antigens were obtained by Multitest IMC (Pasteur Merieux, Lyon, France), before and after 12 and 24 months of therapy.
Arithmetic mean change from baseline (±SD) in T-cell subset numbers, log10 HIV-1 RNA load and SI of LPR during the 24 months follow-up of patients were evaluated by Student‚s t test for paired and unpaired data, as appropriate. Polynomial regression analysis was used to correlate HIV-1 RNA levels and CD4 T-cell count increase from baseline after 24 months.
All patients reached the 24th month of follow-up, regardless of the initial treatment assigned. The HAART regimen consisted of one protease inhibitor (indinavir in 15 patients and ritonavir in six) plus two nucleoside reverse transcriptase inhibitors (zidovudine + didanosine, zidovudine + lamivudine, stavudine + didanosine and lamivudine + stavudine). Constitutional symptoms, when present, resolved shortly after starting HAART. Relapses of opportunistic infections occurred only in two patients in the first 2 months of therapy (cytomegalovirus retinitis and cryptococcal meningitis). No major protease inhibitor-related toxicities were observed in this group. The majority of patients experienced a gain in body weight from baseline values in the first 6 months of therapy (mean increase 2.8±0.6kg), which persisted during follow-up. Karnofsky‚s score increased in all subjects from a mean of 74.7 at baseline to a mean of 91.4 at month 24. Three patients with cytomegalovirus disease permanently discontinued gancyclovir maintenance therapy after 12 months of HAART, remaining free of relapses. Another four patients stopped Pneumocystis carinii pneumonia prophylaxis. In one patient Kaposi‚s sarcoma skin lesions disappeared without other treatment.
Delayed type hypersensitivity
Delayed type hypersensitivity against seven antigens (PPD, streptokinase-streptodornase, tetanus, diphteria, Candida, Proteus and Trichophyton) was negative in all subjects before starting therapy. After 24 months skin test reactivity against at least one antigen was present in four patients showing a mild infiltrate, whereas in other eight subjects an erythematous reaction was observed.
Lymphocyte phenotyping and plasma HIV-RNA
Table 1 shows mean values (ranges) ±SD of T-cell subsets. A significant increase in CD4 T cells was observed from baseline to month 4 (from 20±15¥106/l at entry, to 162±83¥106/l at month 4; P=0.0009). This rise continued during follow-up (334±107¥106/l 24 months later; P<0.0001 versus baseline); however, the number of CD4 cells remained far below normal values. CD8 T cells also showed a significant and consistent increase during the 2 years of follow-up (from 457±327¥106/l at entry, to 1153±589¥106/l at month 24; P=0.0002). Analysis of HLA-DR expression on CD8 T cells did not show a significant variation during the follow-up, despite a trend towards a decline after 24 months (from 171±169¥106/l at baseline, to 107±106¥106/l 24 months later; P=0.3, non-significant). CD8 CD28 T cells showed a significant increase at month 4 (from 111±49¥106/l at entry, to 291±87¥106/l at month 4; P=0.0001). Thereafter CD8 CD28 T cells continued to increase, reaching 443±205¥106/l 24 months after starting HAART (P<0.0001 from baseline values).
A significant mean reduction in HIV-RNA plasma load (>1.5log10 copies/ml) was observed in the whole group from month 4 and was mantained during the follow-up period (from 5.0±0.5 to 3.3±1.0 at month 4, 3.1±1.1 at month 12 and 3.3±1.1 at month 24; P<0.0001 at each time-point versus baseline). Six patients reached undetectable levels of plasma viremia (below 200 copies/ml) at month 4, nine patients reached this at months 12 and 24, whereas in another four patients HIV-RNA fell below 400 copies/ml at months 4 and 12. In 14 patients the decrease of viral load reached 2log10 copies/ml at month 24 compared with baseline levels. In seven cases HIV-RNA persisted unchanged or showed a small decrease (<0.5log10); these were considered to be non-responders, at least in terms of viral burden reduction. Despite the persistent high HIV-RNA levels, these patients increased CD4 T-cell counts from 28±17¥106/l at entry, to 291±150¥106/l at month 24. However, patients who consistently had plasma HIV-RNA levels below limits of detection showed an increase of circulating CD4 T cells significantly higher at month 24 (400±65 versus 301±111¥106/l; P<0.05). Moreover, regression analysis of CD4 increase versus HIV-RNA levels at month 24 showed a significant inverse relationship (r, 0.52, P=0.015, n=21). Mean CD8 T-cell count was higher, although not significantly so, in non-responder patients at month 24 in comparison with the virological responder subjects (1415±736 versus 1013±462¥106/l). CD28 expression on CD8 T cells was the same in the two groups of patients after 24 months of HAART, whereas CD8DR T cells showed a more pronounced decrease in virological responder subjects (78±63 versus 170±150¥106/l).
Memory and naive CD4 T-cell subsets
Analysis of CD45R0 and CD45RA expression on CD4 T cells showed that memory cells (CD45R0) mainly contributed to the initial increase of CD4 T cells. The greater rise was evident in the first 4 months of therapy (from 11±12¥106/l at entry to 130±77¥106/l at month 4; P=0.001), whereas at 12 and 24 months a smaller increase of CD45R0 cells was observed (155±56¥106/l and 160±65¥106/l, respectively). The rise of naive (CD45RA) CD4 T cells was more pronounced between months 12 and 24 (133±61¥106/l; P<0.0001 versus baseline) also if an initial small, but statistically significant, increase was observed. However, their number remained below normal range, reflecting the predominance of the memory subset in the CD4 T-lymphocyte expansion (Fig. 1). The increase in memory and naive CD4 T-cell subsets was similar in virological responder and in non-responder subjects (data not shown).
LPR to mitogens
Before starting HAART all patients showed a defective LPR to anti-CD3 MAb (mean SI, 10.70±2.77; range, 6.58-14.84), but after 24 months of treatment in all but one patient a recovery of anti-CD3 driven lymphoproliferation was observed, up to normal values (SI, 52.14±29.97; range, 9.04-116.30). This significant (P<0.0001) increase of LPR to anti-CD3 was associated with a significant increase of phytohemagglutinin-driven LPR compared with baseline values (SI, 25.65±15.54; range, 10.45-72.28 before HAART and SI, 105.93±52.50; range, 22.96-216.68 after 24 months of treatment; P<0.0001). The patient who did not show recovery of LPR to anti-CD3 after HAART did not restore lymphoproliferation to phytohemagglutinin either (from 12.80 to 22.96). LPR to mitogens are shown in Fig. 2a. LPR to anti-CD3 did not differ in patients with persistent high viral load in comparison with subjects with a significant decrease in plasma HIV-RNA levels after HAART.
LPR to recall and HIV-1 antigens
Before treatment only two patients showed a weak response to mannoprotein (SI, 5.01 and 6.39) and none responded to tetanus toxoid or rgp160. After 24 months of HAART two other patients showed a SI>5 (5.21 and 5.34). The proportion of responder patients who showed positive LPR to mannoprotein increased from 9.5 to 19%. After 24 months of study only one patient responded to rgp160 (SI, 2.62), and none responded to tetanus toxoid. LPR to recall and HIV-1 antigens are shown in Fig. 2b.
Response to HAART can be demonstrated by several clinical parameters such as absence or disappearance of recurrent opportunistic infections, increase of body weight and Karnofsky‚s score increase. Immune restoration may be revealed by quantitative analysis of lymphocyte subpopulations and recovery of responses to recall or immunization antigens [5-7,14].
In this study we have provided an analysis of T-cell subset changes and immune functions in 21 drug- experienced and severely immunocompromised HIV-1 patients during 24 months of HAART. The improvement in Karnofsky‚s index, the body weight increase and the decrease of opportunistic infections observed in our cohort are clinical signs indirectly pointing to immunological recovery.An early increase in memory CD4 T lymphocytes was observed shortly after starting HAART, probably due to a de-trapping of cells from lymphoid tissue, as hypothesized previously [15-17]. Naive CD4 T-cells also showed an increase; however, in our patients this increase was observed later than in less immunocompromised patients studied by others [5,7,14] as it consistently occurred only after 12 months of therapy. The comparison between CD4 and CD8 T-cell increase after HAART is clearly in favour of CD4 T cells (an increase of almost 17 times the baseline level) as opposed to CD8 T cells (an increase of 2.5 times the baseline level). This seems to contrast with the hypothesis of blind T-cell homeostasis , and may be explained by a strong CD4 T-cell proliferation  and perhaps also by thymic differentiation [14,20]. This latter possibility may be supported by the regeneration of naive cells and also by the improvement in CD4 T-cell receptor repertoire observed by others . Alternatively, a decrease in decay rate may be different in CD4 and CD8 T cells after HAART. In previous reports other authors described a subsequent decrease of CD8 T cells [5,14]. We did not observe this decrease perhaps due to the different characteristics of the patients. The expansion of CD8 T cells is accompanied by the decreased expression of activation markers and an increase of CD28 expression, and it is probably due to a reduction of spontaneous cell death . However after 4 months this expansion wanes, perhaps because of the reduction of CD8 stimulation by infected cells due to viral load decrease .
The decline of HIV-RNA levels in virological responder patients, even though partial, seems to represent the most important event caused by HAART. However these subjects were all treated previously with one or two nucleoside reverse transcriptase inhibitors and therefore the possibility that they were partially resistant to these drugs cannot be ruled out. Seven patients who did not show a HIV-RNA decrease had the same clinical improvement such as weight increase and regression of opportunistic infections, increased their naive and memory CD4 T-cell counts and also had a partial functional recovery of LPR. Similar results were recently observed by others . This phenomenon is difficult to explain, but the following hypotheses can be proposed: (i) possible protease inhibitor action on non-viral targets participating in the mechanisms of CD4 T-cell depletion; (ii) presence of low cytopathic or defective viral strains; (iii) transient or partial reduction of viral load (less than 0.5log10) may have been sufficient to allow cell proliferation or to decrease mortality rate and thus to compensate for CD4 T-cell loss. Further studies will be needed in order to verify these hypotheses.
In the course of HIV infection, viral replication plays a pivotal role [24,25]. However, approximately 20% of patients rapidly progress to AIDS even in the presence of low virus replication and, conversely, a proportion of cases (10-25%) with high viral load do not show rapid progression . These facts demonstrate that the immune system is subjected to a balance regulated by the regenerative capacity of progenitor cells, the different cellular mortality rate and the average life span of CD4 and CD8 T lymphocytes. Therefore, although viral replication is the main cause of impairment of immune responses, it represents only one aspect of the balance.
The partial immunological recovery obtained in our patients is only shown by functional tests such as LPR to phytohemagglutinin and anti-CD3 MAb. Responses to mannoprotein and tetanus toxoid were largely unmodified after 24 months of therapy. This could be explained by considering that some recall antigens without antigen stimulation (i.e. tetanus toxoid) could lack memory cell function whereas the naive cells without antigen contact are blind. Only tetanus toxoid booster administration or immunization with new antigens (i.e. haemocyanin), if ethical, could assess the extent of immune reconstitution by priming the newly produced naive T cells. However in their paper, Li et al.  observed recovery of responses to cytomegalovirus and PPD in some patients. Probably these are ubiquitous antigens in immunocompromised patients and they were able to stimulate lymphocytes. Nevertheless, we did not observe an immune response to Candida albicans antigen, which is also a ubiquitous microorganism. We can not exclude that in a prolonged follow-up restoration of response to recall antigens may be observed. The response to HIV-1 rgp160 was not restored as in the patients described by Pontesilli et al. . The diminution of viral antigens following HAART may not help CD4 and CD8 T-cells to mount an HIV-specific response. However the seven patients with increased CD4 T cells and high viral load also failed to present a specific T-cell response to gp160. Other studies are needed to clarify the mechanism of immunologic correction and the possible stimulation with specific antigens in vivo, possibly in conjunction with adjuvants. We  and others  have demonstrated previously that the response to specific viral antigens can be obtained only by active immunization.
In conclusion, our data demonstrate the possibility of a partial restoration in T-cell subsets and immune functions also in patients starting HAART at an advanced stage of HIV-1 disease. The clinical benefits and the decrease of opportunistic infections observed also suggest that our immunological tests are not able, at present, to indicate precisely the degree of immune recovery. In some patients the immunological and clinical recovery may be independent of the virological outcome and suggest that even a partial reduction in viral load may be beneficial, at least in this group of patients.
Combined treatments of HAART, cytokines (interleukin-2, interleukin-12, etc.) and vaccines (proteins and peptides) may be considered for a strategy to obtain full immune reconstitution in HIV-1-infected patients.
1. Giorgi JV, Detels R. T-cell subset alterations in HIV-infected homosexual men: NIAID Multicenter AIDS Cohort study.Clin Immunol Immunopathol
2. Clerici M, Stocks NI, Zajac RA, et al
. Detection of three distinct patterns of T helper cell dysfunction in asymptomatic, human immunodeficiency virus-seropositive patients.J Clin Invest
3. Hammer SM, Squires KE, Hughes MD, et al
. A controlled trial of two nucleoside analogues plus indinavir in persons with human immunodeficiency virus infection and CD4 cell counts of 200 per cubic millimeter or less.N Engl J Med
4. Gulick RM, Mellors JW, Havlir D, et al
. Treatment with indinavir, zidovudine, and lamivudine in adults with human immunodeficiency virus infection and prior antiretroviral therapy.N Engl J Med
5. Autran B, Carcelain G, Li TS, et al
. Positive effects of combined antiretroviral therapy on CD4+ T cell homeostasis and function in advanced HIV disease.Science
6. Emery S, Lane HC. Immune reconstitution in HIV infection.Curr Opin Immunol
7. Li TS, Tubiana R, Katlama C, Calvez V, Ait Mohand H, Autran B. Long-lasting recovery in CD4 T-cell function and viral-load reduction after highly active antiretroviral therapy in advanced HIV-1 disease.Lancet
8. Connors M, Kovacs JA, Krevat S, et al
. HIV infection induces changes in CD4+ T-cell phenotype and depletions within the CD4+ T-cell repertoire that are not immediately restored by antiviral or immune-based therapies.Nature Med
9. Gray CM, Schapiro JM, Winters MA, Merigan TC. Changes in CD4+ and CD8+ T cell subset in response to highly active antiretroviral therapy in HIV type 1-infected patients with prior protease inhibitor experience. AIDS Res Hum Retroviruses
10. Centers for Disease Control. 1993 revised classification system for HIV infection and expanded surveillance case definition for AIDS among adolescents and adults.MMWR
11. Pontesilli O, Guerra EC, Ammassari A, et al
. Phase II controlled trial of post-exposure immunization with recombinant gp160 versus antiretroviral therapy in asymptomatic HIV-1-infected adults. VaxSyn Protocol Team.AIDS
12. Quinti I, Palma C, Guerra EC et al
. Proliferative and cytotoxic responses to mannoproteins of Candida albicans by peripheral blood lymphocytes of HIV-infected patients. Clin Exp Immunol
13. Schrier RD, Gnann JW, Landes R, et al
. T cell recognition of HIV synthetic peptides in a natural infection.J Immunol
14. Pakker NG, Notermans DW, de Boer RJ, et al
. Biphasic kinetics of peripheral blood T cells after triple combination therapy in HIV-1 infection: A composite of redistribution and proliferation. Nature Med
15. Mosier D. HIV results in the frame. CD4+ cell turnover.Nature
16. Sprent J, Tough D. HIV results in the frame. CD4+ cell turnover. Nature
17. Roederer M. Getting to the HAART of T cell dynamics.Nature Med
18. Margolick JB, Munoz A, Donnenberg AD, et al
. Failure of T-cell homeostasis preceding AIDS in HIV-1 infection. The Multicenter AIDS Cohort Study.Nature Med
19. Walker RE, Carter CS, Muul L, et al
. Peripheral expansion of pre-existing mature T cells is an important means of CD4+ T-cell regeneration HIV-infected adults.Nature Med
20. Hellerstein M, Hanley MB, Cesar D, et al
. Directly measured kinetics of circulating T lymphocytes in normal and HIV-1 infected humans.Nature Med
21. Gorochow G, Neuman AU, Kereveur A, et al
. Perturbation of CD4+ and CD8+ T-cell repertoires during progression to AIDS and regulation of the CD4+ repertoire during antiviral therapy.Nature Med
22. Pandolfi F, Pierdominici M, Marziali M, et al
. Modification of apoptosis during combination therapy. 12th World AIDS Conference
. Geneva, June 1998 [abstract 42184].
23. Levy JA. Caution: should we be treating HIV infection early?Lancet
24. Mellors JW, Kingsley LA, Rinaldo CR, et al
. Quantitation of HIV-1 RNA in plasma predicts outcome after seroconversion.Ann Intern Med
25. Mellors JW, Munoz A, Giorgi JV, et al
. Plasma viral load and CD4+ lymphocytes as prognostic markers of HIV-1 infection.Ann Intern Med
26. Pontesilli O, Kerkhof-Garde S, Pakker NG, et al
. Functional lymphocyte reconstitution and HIV-1 specific T cell responses during highly active antiretroviral therapy (HAART).12th World AIDS Conference.
Geneva, June 1998 [abstract 31134].
27. Carlesimo M, Pontesilli O, Guerra EC, et al
. Long-term evaluation of cellular immunity during antiretroviral therapy and immunization with human immunodeficiency virus type 1 (HIV-1) env glycoprotein in HIV-1-infected persons.J Infect Dis
28. Moss RB, Trauger JR, Giermakowska WK, et al
. Effect of immunization with an inactivated gp120 depleted HIV-1 immunogen (REMUNETM) on b chemokine and cytokine production in subjects with HIV-1 infection. J Acquir Immune Defic Syndr Hum Retrovirol